DOCSLIB.ORG

Underspecification of Intersective Modifier

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Underspecification of Intersective Modifier Underspecification of Intersective Modifier Attachment: Some Arguments from German Berthold Crysmann DFKI, Saarland University Proceedings of the 11th International Conference on Head-Driven Phrase Structure Grammar Center for Computational Linguistics, Katholieke Universiteit Leuven Stefan Muller¨ (Editor) 2004 CSLI Publications pages 378–392 http://csli-publications.stanford.edu/HPSG/2004 Crysmann, Berthold. 2004. Underspecification of Intersective Modifier Attach- ment: Some Arguments from German. In Muller,¨ Stefan (Ed.), Proceedings of the 11th International Conference on Head-Driven Phrase Structure Grammar, Center for Computational Linguistics, Katholieke Universiteit Leuven, 378–392. Stanford, CA: CSLI Publications. Abstract In this paper, I shall discuss the semantic attachment of intersective mod- ifiers in German coherent constructions. I shall show that a purely syntactic solution to the observable attachment ambiguity is undesirable for reasons of processing efficiency and/or massive spurious ambiguity. Instead, I shall follow Egg and Lebeth (1995) and propose an extension to Minimal Recur- sion Semantics, permitting the expression of underspecified semantic attach- ment. This rather trivial move, as we shall see, will not only be preferable for processing reasons, but it will also be more in line with the spirit of under- specified semantics, effectively providing a compact representation of purely semantic distinctions, instead of unfolding these distinctions into a rain forest of tree representations and derivations. I will present an implementation of the underspecification approach integrated into the German HPSG developed at DFKI and compare its efficiency to an alternative implementation where semantic attachment is unfolded by means of retrieval rules. 1 Intersective modifiers and word order It is a well-known property of German that order in the Mittelfeld is extremely free: although some restrictions do seem to exist as to the relative order in which a verb’s complements can appear, it is by now generally accepted that the linearisa- tion constraints regulating order within the Mittelfeld should best be conceived of as soft constraints or performance preferences (Uszkoreit, 1987). The word-order freeness of German is further multiplied by the fact that auxiliaries, modals, con- trol and raising verbs may or must construct coherently (Kiss, 1994, 1995, Muller,¨ 1999, Hinrichs and Nakazawa, 1990), a construction that is modelled by means of Hinrichs/Nakazawa-style argument composition. What is more, inherent and in- herited arguments can, again, undergo scrambling, thus, in principle, arguments of the upper and lower verbs may appear in any order. One of the fundamental empirical tests for the coherent construction — besides scrambling of arguments, of course — builds on the interpretation of modifiers. With a few exceptions, e.g. the marker of sentential negation nicht ‘not’, there does not appear to be any general positional restriction on the distribution of modi- fiers in the Mittelfeld: as a rule of thumb, modifiers can appear just about anywhere between the left and right sentence bracket, demarcated by a complementiser or a finite verb and the sentence-final verb cluster. Independently of position, how- ever, modifiers in the coherent construction often display a systematic ambiguity between high and low attachment. †This work has been carried out as part of the DFKI project QUETAL, funded by the German Federal Ministry for Education and Research (BMBF), and the EU-project DEEPTHOUGHT at the Department for Computational Linguistics, Saarland University. The ideas presented in this paper have benefited a lot from discussion with various people. I would therefore like to thank Ann Copestake, Markus Egg, Dan Flickinger, Stefan Muller,¨ Stephan Oepen and Emily Bender for their comments and suggestions. 379 (1) Peter hat es im Labor blitzen sehen Peter has it in.the lab flash saw ‘Peter saw some flashes/lightning in the lab’ (P”utz, 1982, 340) As exemplified by the datum above, the PP im Labor can modify either the seeing event, or the flashing event: under the first interpretation, Peter is in the lab observing some lightning or flashes somewhere else (possibly outside), whereas under the latter, the flashes are in the lab, with the locus of the observer unspecified. Although, in (1) the modifier is adjacent to the verb cluster, permitting us to model the two semantic interpretations by means of high or low syntactic attach- ment, this is not always the case: as illustrated in (2), a flipped auxiliary may intervene (Hinrichs and Nakazawa, 1994, Kathol, 2000, Meurers, 2001), making adjunction to the most deeply embedded cluster element impossible.1 Still, the modifier displays the same semantic attachment ambiguity as in the example in (1) above. (2) weil Peter es im Labor [hat [[blitzen] sehen]] because Peter it in.the lab has flash saw ‘because Peter saw some flashes/lightning in the lab’ The very same can be observed with scrambling in the Mittelfeld: (3) a. weil Peter im Labor es blitzen sah because Peter in.the lab it flash saw ‘because Peter saw some flashes/lightning in the lab’ b. weil im Labor Peter es blitzen sah because in.the lab Peter it flash saw ‘because Peter saw some flashes/lightning in the lab’ Independent of surface position, and, therefore, constituency, modification of upstairs and downstairs verb is equally possible. Similar evidence against a purely syntactic approach to intersective modifier attachment is provided by Egg and Lebeth (1995): (4) Sollen wir im Marz¨ noch einen Termin machen? shall we in March an appointment make ‘Should we schedule a meeting in March?’ (Egg and Lebeth, 1995) The sentence in (4) is three-ways ambiguous: the PP adjunct im M¨arz ‘in March’ may modify the appointment (Termin), the scheduling event (ausmachen), or even the modal (sollen). Under standard assumptions of phrase structure in 1For the purposes of this paper I will concentrate foremost on versions of HPSG without word order domains. As far as I can tell, the issues raised within the scope of this paper are by-and-large the same for linearisation approaches and true movement analyses (see below). 380 the German modal constructions (Kiss, 1994), only attachment to sollen should be available. Attachment to the main verb infinitive, however, can only be derived by making otherwise unmotivated assumptions about phrase structure, namely that modals optionally take a VP constituent as their complement. It should be clear that the data presented thus far constitute a syntax-semantics mismatch: ceteris paribus, modification of the downstairs verb obviously conflicts with straightforward rule-by-rule compositionality. Thus, some more elaborated mechanisms are called for to derive the full set of interpretations, independent of constituency in the Mittelfeld. 1.1 Storage and Retrieval One such extension has been proposed in Kiss (1995): to overcome the kind of problem just sketched, he proposes to collect modification targets in a special stor- age feature from which they can be retrieved whenever a modifier is attached in syntax. Introduction of modification targets onto the storage works in tandem with verb complex formation. Though certainly a viable solution at the time, nowadays, such an approach is not anymore fully attractive, with the Cooper-storage being successfully supplanted by much more concise underspecified descriptions. Fur- thermore, in a computational setting,2 retrieval during parsing can be quite costly, as the exact number of modification targets is locally not always known in bottom- up parsing. Owing to the fact that entire verb clusters can be extracted into the Vorfeld, the complexity of the extracted cluster is unknown at the point where the Mittelfeld is constructed. Thus, whenever Partial VP Fronting can be hypothesised during parsing, the number of available modification targets to be assumed locally will be equal to the maximum complexity of verbal clusters in German. (5) Blitzen sehen hat Peter es im Labor. flash see has Peter it in.the lab ‘Peter saw some flashes/lightning in the lab’ Even if we can put an upper bound on verb cluster complexity — the most complex cluster I found in Meurers (1997) consisted of 5 elements in total —, it should be kept in mind that retrieval of modification targets during parsing will increase by this factor not only the number of head-modifier edges themselves but also the number of chart items that can be transitively derived from these edges. Although the overall frequency of partial VP fronting in German is not that high, local ambiguity is unaffected by this, due to the unbounded nature of the process: even the chart of “harmless” sentences without any PVP fronting is characterised by an incommensurate number of PVP hypotheses. 2As a point of reference I use the fastest processing platform for HPSG grammars currently available, namely PET (Callmeier, 2000) together with the development platform LKB (Copestake, 2001). As for the grammar, I will assume the large-scale grammar of German, developed in the Verbmobil context by Muller¨ and Kasper (2000), which has been ported to LKB/PET by Stefan Muller¨ and subsequently enhanced by Berthold Crysmann (Crysmann, 2003, to appear). 381 Based on these two objections, we can discard the storage-retrieval approach as a suboptimal solution, at least, unless a more efficient and elegant solution can be found. 1.2 Scrambling as movement Another obvious way to attack the issue is to analyse scrambling as movement, akin to analyses carried out in the generative paradigm. Besides the issue whether or not one should treat essentially local order phenomena on a par with unbounded dependencies, an extraction-based approach will introduce a fair amount of spu- rious ambiguity into the grammar: unless we can canonicalise the introduction of modifier gaps in a highly restrictive fashion, regulating the relative attachment for every pair of different modifiers, even simple sentences with only one modification target but two intersective modifiers will end up with two syntactically different, yet semantically identical analyses. Worse, the amount of spurious ambiguity thus introduced will be factorial to the number of modifiers present.
Load more

Recommended publications
  • The Generative Lexicon
    The Generative Lexicon James Pustejovsky" Computer Science Department Brandeis University In this paper, I will discuss four major topics relating to current research in lexical seman- tics: methodology, descriptive coverage, adequacy of the representation, and the computational usefulness of representations. In addressing these issues, I will discuss what I think are some of the central problems facing the lexical semantics community, and suggest ways of best ap- proaching these issues. Then, I will provide a method for the decomposition of lexical categories and outline a theory of lexical semantics embodying a notion of cocompositionality and type coercion, as well as several levels of semantic description, where the semantic load is spread more evenly throughout the lexicon. I argue that lexical decomposition is possible if it is per- formed generatively. Rather than assuming a fixed set of primitives, I will assume a fixed number of generative devices that can be seen as constructing semantic expressions. I develop a theory of Qualia Structure, a representation language for lexical items, which renders much lexical ambiguity in the lexicon unnecessary, while still explaining the systematic polysemy that words carry. Finally, I discuss how individual lexical structures can be integrated into the larger lexical knowledge base through a theory of lexical inheritance. This provides us with the necessary principles of global organization for the lexicon, enabling us to fully integrate our natural language lexicon into a conceptual whole. 1. Introduction I believe we have reached an interesting turning point in research, where linguistic studies can be informed by computational tools for lexicology as well as an appre- ciation of the computational complexity of large lexical databases.
    [Show full text]
  • Property Modifiers
    Property Modifiers Marie Duží and Michal Fait VSB-Technical University Ostrava, Department of Computer Science FEI, 17. listopadu 15, 708 33 Ostrava, Czech Republic [email protected], [email protected] Abstract. In this paper, we deal with property modifiers defined as func- tions that associate a given root property P with a modified property [MP]. Property modifiers typically divide into four kinds, namely intersective, subsective, privative and modal. Here we do not deal with modal modi- fiers like alleged, which appear to be well-nigh logically lawless, because, for instance, an alleged assassin is or is not an assassin. The goal of this paper is to logically define the three remaining kinds of modifiers. Fur- thermore, we introduce the rule of pseudo-detachment as the rule of left subsectivity to replace the modifier M in the premise by the property M* in the conclusion, and prove that this rule is valid for all kinds of modifiers. Furthermore, it is defined in a way that avoids paradoxes like that a small elephant is smaller than a large mouse. Key words: Property modifier, subsective, intersective, privative, the rule of pseudo-detachment, Transparent intensional logic, TIL, intensional essentialism 1 Introduction We introduce a logic of property modifiers modelled as a mapping from proper- ties to properties, such that the result of the application of a modifier to a property is another property. This is because the result of modification does not depend on the state of the world, nor on time. For instance, if one applies the modifier Skilful to the property Surgeon, they obtain the property of being a skilful surgeon.
    [Show full text]
  • Intersective Entailments with Adverbials?
    Class 4 Modification and Events Modifiers in syntax Modifier = constituent that doesn’t change syntactic category of its sister. happy – adjective (phrase) very happy – adjective phrase (AP) very is an AP modifier women – noun (phrase) tall women – noun phrase (NP) tall is an NP modifier (adnominal) More examples? ran quickly – verb phrase (VP) modifier (adverb, adverbial) very quickly – (also) adverb modifier Further? men in the kitchen, men who knit, men and/or women, almost every man Semantics: Modifier Functions Modifier function = a function of type aa. happy – et very happy – et very – (et )( et ) very denotes a modifier function women – et tall women – et tall – (et )( et ) every man – (et )t Alternatively: almost every man – (et )t every – (et )(( et )t) almost – (( et )t)(( et )t) almost every – (et )(( et )t) almost – (( et )(( et )t))(( et )(( et )t)) Syntax-Semantics Hypothesis: All modifiers in syntax denote modifier functions. [ In categorial grammar: Type(N)=et Type(N N)=(et)(et) Type(NP)=(et)t Type(NP NP)=((et)t)((et)t) In general: Type(X X) = (Type(X))(Type(X)) ] But what denotations should modifier functions have? Intersective adnominal modifiers [[ fast cars ]] = car ∩ fast [[ houses in England ]] = house ∩ in_England [[ houses where I lived ]] = house ∩ { x∈E : live_in(I,x) } A function F of type ( et )( et ) is an intersective modifier if there is a set of entities B s.t. for every function get characterizing a set A, the et function F(g) characterizes the set A∩B. Adverbials – similarity to adnominals Adnominals Adverbials Adjectives/ Adverbs fast cars ate fast Preposition Phrases houses in England ate in England Relative Clauses houses where I lived ate where I lived Question: Can we treat adverbials as intersective modifiers? Common answer: - Surely not as (et)(et) modifiers.
    [Show full text]
  • Redalyc.Property Modifiers and Intensional Essentialism
    Computación y Sistemas ISSN: 1405-5546 [email protected] Instituto Politécnico Nacional México Duží, Marie Property Modifiers and Intensional Essentialism Computación y Sistemas, vol. 21, núm. 4, 2017, pp. 601-613 Instituto Politécnico Nacional Distrito Federal, México Available in: http://www.redalyc.org/articulo.oa?id=61553900004 How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative ISSN 2007-9737 Property Modifiers and Intensional Essentialism Marie Duží VSB-Technical University of Ostrava, Department of Computer Science, Czech Republic [email protected] Abstract. In this paper, I deal with property modifiers characteristic function from individuals to truth- defined as functions that associate a given root property values. So, for an individual to instantiate a P with a modified property [M P]. Property modifiers property, whether modified or not, is to be an typically divide into four kinds, namely intersective, element of its extension at a given world and time. subsective, privative and modal. Here I do not deal with The novel contribution of this paper is a new modal modifiers like alleged, which appear to be well- nigh logically lawless, because, for instance, an alleged definition of subsective and privative modifiers in assassin is or is not an assassin. The goal of this paper terms of intensional essentialism. is to logically define the three remaining kinds of Kamp’s seminal [9] seeks to draw a line modifiers together with the rule of left subsectivity that I between those adjectives whose meaning is a launch as the rule of pseudo-detachment to replace the property and those adjectives whose meaning is a modifier M in the premise by the property M* in the function that maps properties to properties.
    [Show full text]
  • Lecture 8. Semantics of Possessives, Continued. Arguments Vs. Modifiers
    The Structure of Meaning, Lecture 8 Barbara H. Partee, March 20, 2006 Lecture 8. Semantics of Possessives, continued. Arguments vs. modifiers. Puzzles of predicate possessives. 1. Semantics of possessives, continued: arguments vs. modifiers. ................................................................................1 2. Different kinds of genitives/possessives: why Jensen and Vikner may be right for Russian genitives but not for all English genitives. ..........................................................................................................................................3 2.1. Predicate possessives: a problem for the “one genitive” approach? ...................................................................4 2.1.1. Predicate possessives in English. ...........................................................................................................5 2.1.2. Russian prenominal possessives vs. genitives........................................................................................6 2.1.3. German possessive pronouns. ................................................................................................................6 2.1.4. Polish possessive pronouns....................................................................................................................8 2.2. Conclusions about predicate possessives. ...........................................................................................................9 REFERENCES ..............................................................................................................................................................9
    [Show full text]
  • Correlative Clause Features in Sanskrit and Hindi/Urdu
    CORRELATIVE CLAUSE FEATURES IN SANSKRIT AND HINDI/URDU Alice Davison, University of Iowa [email protected] 12-10-06 1. Introduction Correlative clauses represent a parametric variation on relative clauses, found in various related and unrelated languages (cf. Grosu 2002, den Dikken 2005) In this paper I explore how this parameter is realized over a period of some thousands of years in Indic languages. I contrast correlative clauses related finite clauses in the earliest Indic language which is attested, the Sanskrit of the Rg Veda and early Sanskrit prose, with corresponding subordinate clauses in a modern Indic language, Hindi/Urdu. There is remarkable lexical continuity, in that the relative determiners are formally distinct from the interrogatives. Sanskrit has only one dependent clause type, the correlative construction, which corresponds to three kinds of subordinate clause in Hindi/Urdu: correlative clauses, complement clauses, and conditional/adverbial clauses. This comparison allows some exploration of how languages divided in time share a specific parameter (Gianollo et al, to appear), and to what degree they diverge. The two language have many common lexical and structural properties. But by many syntactic and semantic criteria, the correlative clauses in the two languages are sharply different. In Vedic Sanskrit, correlative clauses are loosely and paratactically related to another clause, while in Hindi/Urdu, the relation between a correlative clause and the other ‘host’ clause is very closely constrained, and dependent clauses are syntactically different from main clauses. I propose some formal features which form links between adjoined clauses, and guide semantic interpretation. This sequence of historical changes, which took place at some point between Vedic Sanskrit and the modern languages, involves the grammaticization of a semantic predicational feature, so that what was a default feature becomes a lexical feature of relative Ds.
    [Show full text]
  • Person Features: Deriving the Inventory of Persons
    Chapter 2 Person Features: Deriving the Inventory of Persons 1. Introduction One long-standing aim of research into person has been to achieve a deeper understanding of the inventory of personal pronouns by decomposing them into a limited set of person features. The main aim of this chapter is to make a contribution to this enterprise. We will argue that a set of two privative person features is sufficient to derive the full inventory of attested pronouns and their interpretations, without generating non-attested pronouns. In chapter 1 we showed that the cross-linguistic inventory of attested persons is a small subset of the set of theoretically possible persons. The theoretically possible persons are those that can be formed by freely combining the speaker (for which we use the symbol i), associates of the speaker (ai), the addressee (u), associates of the addressee (au), and others (o). The generalizations that distinguish between the attested and non-attested persons are the following (repeated from (9) in chapter 1): (1) a. There is no person that contains associates of a participant but not the participant itself. b. There is no person that consists of others and only speaker(s) or of others and only addressee(s). c. No person system distinguishes pluralities containing a participant but not its associates from pluralities containing that participant as well as its associates. d. First and second person plural pronouns show an ‘associative effect’: any element contained in them other than i or u must be ai or au and cannot be o. 31 This leaves us with three persons with a singular interpretation and four persons with a plural interpretation.
    [Show full text]
  • Probing Syntax and Semantics with French Propre A
    UNIVERSITY OF CALIFORNIA Los Angeles On her own Parsimonious Compositionality: Probing Syntax and Semantics with French propre A dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy in Linguistics by Isabelle Charnavel 2012 © Copyright by Isabelle Charnavel 2012 ABSTRACT OF THE DISSERTATION On her own Parsimonious Compositionality: Probing Syntax and Semantics with French propre by Isabelle Charnavel Doctor of Philosophy in Linguistics University of California, Los Angeles, 2012 Professor Dominique Sportiche, Chair This dissertation focuses on the French word propre roughly meaning ‘characteristic-of’ and corresponding to English own found in ‘her own thesis’. This adjective makes extremely varied and complex contributions to the meaning and properties of sentences it occurs in. The present work addresses the question of how these contributions arise. Parsimoniously assuming a unique lexical entry for propre, these contributions are compositionally derived by a specific DP-internal structure and different interactions with focus. More precisely, propre is analyzed as taking as argument a possessive relation characterized as most specific. Unlike postnominal propre , prenominal propre exhibits ii three main readings called restrictive, possessor and possessum propre : restrictive propre has a standard intersective truth-conditional effect; possessor and possessum propre do not, but induce focus alternatives respectively to the possessor and to the possessum; possessum propre moreover gives rise to scalarity effects. These readings are argued to derive from a principle of minimization and different interactions with focus; in particular, the behavior of possessum propre shows the presence of a covert focus operator akin to even . When combined with a pronominal possessor like son (‘his’), the behavior of propre provides probes bearing on binding theoretic issues.
    [Show full text]
  • Cross-Linguistic Representations of Numerals and Number Marking∗
    Proceedings of SALT 20: 582–598, 2011 Cross-linguistic representations of numerals and number marking∗ Alan Bale Michaël Gagnon Concordia University University of Maryland Hrayr Khanjian Massachusetts Institute of Technology Abstract Inspired by Partee(2010), this paper defends a broad thesis that all mod- ifiers, including numeral modifiers, are restrictive in the sense that they can only restrict the denotation of the NP or VP they modify. However, the paper concentrates more narrowly on numeral modification, demonstrating that the evidence that moti- vated Ionin & Matushansky(2006) to assign non-restrictive, privative interpretations to numerals – assigning them functions that map singular sets to sets containing groups – is in fact consistent with restrictive modification. Ionin & Matushansky’s (2006) argument for this type of interpretation is partly based on the distribution of Turkish numerals which exclusively combine with singular bare nouns. Section 2 demonstrates that Turkish singular bare nouns are not semantically singular, but rather are unspecified for number. Western Armenian has similar characteristics. Building on some of the observations in section2, section3 demonstrates that re- strictive modification can account for three different types of languages with respect to the distribution of numerals and plural nouns: (i) languages where numerals exclusively combine with plural nouns (e.g., English), (ii) languages where they exclusively combine with singular bare nouns (e.g., Turkish), (iii) languages where they optionally combine with either type of noun (e.g., Western Armenian). Ac- counting for these differences crucially involves making a distinction between two kinds of restrictive modification among the numerals: subsective vs. intersective modification. Section3 also discusses why privative interpretations of numerals have trouble accounting for these different language types.
    [Show full text]
  • Intersective and Subsective Adjectives Ling 5205: Semantics
    Outline Attributive uses of adjectives Practice Summary References Intersective and subsective adjectives Ling 5205: Semantics Brian Reese Institute of Linguistics University of Minnesota [email protected] 1 / 32 Outline Attributive uses of adjectives Practice Summary References 1 Attributive uses of adjectives 2 Practice 3 Summary 2 / 32 Outline Attributive uses of adjectives Practice Summary References 1 Attributive uses of adjectives 2 Practice 3 Summary 3 / 32 Outline Attributive uses of adjectives Practice Summary References Modificational adjectives So far we have considered only predicative occurrences of adjectives; but adjectives also occur before nouns in attributive, or modificational, uses. Examples (1) a. Tina is tall. (Predicative) b. Thin is thin. (2) a. Tina is a tall woman. (Modificational) b. The tall engineer visited us. c. I met five tall astronomers. 4 / 32 Outline Attributive uses of adjectives Practice Summary References Modificational adjectives In many cases there are strong semantic relations between modificational and predicative uses of adjectives. (We will focus on the equivalence in 3a.) Examples (3) a. Tina is a Chinese pianist , Tina is Chinese and Tina is a pianist. b. My doctor wears no white shirts , No shirts that my doctor wears are white. c. Dan saw six carnivorous animals , Six animals that Dan saw are carnivorous. 5 / 32 Outline Attributive uses of adjectives Practice Summary References New semantic assumptions Common nouns denote et functions; indefinite article a when used in predicative NPs denotes the identity function on et functions. Examples (4) Tina [ is [ a pianist]] (5) a. pianist = pianistet J K b. a = a(et)(et) = is = λget:g J K (6) (is(a(pianist)))(tina) = pianist(tina) = 1 if tina 2 pianist∗ and 0 otherwise 6 / 32 Outline Attributive uses of adjectives Practice Summary References Modificational adjectives Example (7) Tina [ is [ a [ Chinese pianist ]]] a *type-mismatch* a(et)(et) Chinese pianist chineseet pianistet Problem: We have no way to compose two et functions into an et function using function application.
    [Show full text]
  • Hy-NLI: a Hybrid System for State-Of-The-Art Natural Language Inference
    Hy-NLI: a Hybrid system for state-of-the-art Natural Language Inference Doctoral thesis submitted for the degree of Doctor of Philosophy (Dr.phil.) presented by Aikaterini-Lida Kalouli at the Faculty of Humanities Department of Linguistics Konstanz, 2021 Konstanzer Online-Publikations-System (KOPS) URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-2-1cwr6rmbvoqxv2 Date of the oral examination: March 18th, 2021 First referee: Prof. Dr. Miriam Butt Second referee: Prof. Dr. Maribel Romero Third referee: Prof. Dr. Lauri Karttunen The human spirit must prevail over technology. Albert Einstein Abstract A main characteristic of human language and understanding is our ability to reason about things, i.e., to infer conclusions from given facts. Within the field of Natural Language Processing and Natural Language Understanding, the task of inferring such conclusions has come to be known as Natural Language Inference (NLI) and is currently a popular field of research. NLI is most often formulated as the task of determining where a sentence entails (i.e., implies) or contradicts (i.e., implies the opposite) or is neutral (i.e., does not have any relation) with respect to another sentence (MacCartney and Manning, 2007). Although such a task sounds trivial for humans, it is less so for machines: the processes behind human inference require even more than understanding linguistic input; they presuppose our understanding about the world and everyday life and require the complex combination and interaction of this information. In this thesis, I implement a hybrid NLI system, Hy-NLI, which is able to deter- mine the inference relation between a pair of sentences.
    [Show full text]
  • MODIFICATION? 9 Not Elsewhere
    DRAFT Modification Marcin Morzycki Michigan State University June 12, 2013 Latest revision: June 2014 This is a draft of a book in preparation for the Cambridge University Press series Key Topics in Semantics and Pragmatics. It’s something between a textbook for people with a basic background in semantics and an introduc- tory survey of work on the semantics of adjectives, adverbs, and degrees. For a fuller explanation, see section 1.2. Comments would be extremely helpful, so please don’t hesitate to contact me if you have any, even very minor ones. DRAFT Contents Acknowledgments vii 1 Preliminaries 1 1.1 Two problems .............................. 1 1.2 What this book is and isn’t ..................... 2 1.3 Background assumptions ...................... 3 1.3.1 Glossing logical notation ................. 3 1.3.2 Theoretical framework ................... 4 1.3.3 Notational and typographical conventions ....... 7 1.4 What, if anything, is modification? ................ 8 1.5 Roadmap ................................ 12 2 The Lexical Semantics of Adjectives: More Than Just Scales 13 2.1 Introduction .............................. 13 2.2 How adjectives and nouns combine: a typology . 14 2.2.1 Intersective interpretations . 14 2.2.2 Subsective interpretations . 16 2.2.3 Apparently subsective intersective interpretations . 20 2.2.4 Ordinary non-subsective adjectives . 22 2.2.5 Privative adjectives, which may not exist . 24 2.3 The type of adjectives and the nature of subsectivity . 26 2.3.1 How powerful are adjectives? . 26 2.3.2 Siegel: The Doublet Theory . 30 2.3.3 Larson: events inside the nominal extended projection 34 2.3.4 The implicit argument approach .
    [Show full text]